Discovery of intestinal targeted TGR5 agonists for the treatment of type 2 diabetes

Article abstract of DOI:10.1021/jm500829b

Activation of TGR5 stimulates intestinal glucagon-like peptide-1 (GLP-1) release, but activation of the receptors in gallbladder and heart has been shown to cause severe on-target side effects. A series of low-absorbed TGR5 agonists was prepared by modify

Full text of DOI:10.1021/jm500829b

Subscriber access provided by UNIV LAVAL
Article
Discovery of intestinal targeted TGR5
agonists for the treatment of type 2 diabetes
hongliang duan, mengmeng ning, qingan zou, yangliang
ye, ying feng, lina zhang, Ying Leng, and Jianhua Shen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm500829b • Publication Date (Web): 24 Feb 2015
Downloaded from http://pubs.acs.org on March 4, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of intestinal targeted TGR5 agonists for
the treatment of type 2 diabetes
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hongliang Duan,† Mengmeng Ning,† Qingan Zou, Yangliang Ye, Ying Feng, Lina Zhang, Ying
Leng,* Jianhua Shen*
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica (SIMM), Chinese
Academy of Sciences, Shanghai 201203, China
ABSTRACT
Activation of TGR5 stimulates intestinal glucagonꢀlike peptideꢀ1 (GLPꢀ1) release, but
activation of the receptors in gallbladder and heart has been shown to cause severe on target side
effects. A series of lowꢀabsorbed TGR5 agonists was prepared by modifying compound 2 with
polar functional groups to limit systemic exposure and specifically activate TGR5 in the intestine.
Compound 15c, with a molecular weight of 1401, a PSA value of 223 Å2, and low permeability
on Cacoꢀ2 cells, exhibited satisfactory potency both in vitro and in vivo. Low levels of 15c were
detected in blood, bile, and gallbladder tissue, and gallbladder related side effects were
substantially decreased compared to the absorbed smallꢀmolecule TGR5 agonist 2.
INTRODUCTION
TGR5 (GPBAR1, MꢀBAR, GPCR19) is a G proteinꢀcoupled receptor (GPCR) first identified
as a cellꢀsurface bile acid receptor in 2002.1 Activation of TGR5 promotes secretion of glucagonꢀ
like peptideꢀ1 (GLPꢀ1),2 which activates insulin secretion, suppresses glucagon release as well as
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 44
1
2
slows gastric emptying and suppresses appetite.3,4 Elevated GLPꢀ1 concentrations increase
glucose disposal and decrease blood glucose levels. In addition, TGR5 involves in metabolic
regulation in brown fat and skeletal muscle during energy expenditure.5,6 Since obesity and
weight gain are common complications in type 2 diabetes mellitus (T2DM), TGR5 agonists may
provide new treatments forT2DM and associated metabolic diseases.7,8,9
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Human TGR5 shares 83% amino acid sequence identity with TGR5 in mice.10 TGR5 gene
expression is widely distributed, with moderate expression levels in the intestine, spleen, and
placenta, and greater expression level in the gallbladder.11,12 Activation of TGR5 in gallbladder
epithelium causes smooth muscle relaxation, stimulation of gallbladder filling, and an acute
increase in gallbladder volume.13 TGR5 agonists also produce cardiovascular effects, such as
changes in heart rate.14,15 The gallbladder and cardiac side effects are the main obstacles to the
clinical application of TGR5 agonists.
There are two categories of TGR5 agonists at present. The first series includes naturally
occurring bile acids (BAs) such as cholic acid (CA) and lithocholic acid (LCA),1 and their
semisynthetic derivatives such as 6αꢀethylꢀ23(S)ꢀmethylcholic acid (INT777, 1, Figure 1),16
while the second series consists of synthetic small molecules.17ꢀ23 We previously reported a
series of orally efficacious TGR5 agonists based on 4ꢀphenoxynicotinamide, of which one of the
most potent molecules was [4ꢀ(2,5ꢀdichlorophenoxy)ꢀpyridinꢀ3ꢀyl]ꢀ(4ꢀcyclopropylꢀ3,4ꢀdihydroꢀ
2Hꢀquinoxalinꢀ1ꢀyl)ꢀmethanone (2, Figure 1).24 These agonists exhibited nanomolar potency and
were proved to be effective in lowering glucose levels following an oral administration. High
exposure to intestine caused the secretion of GLPꢀ1, but exposure to other tissues such as
gallbladder and heart resulted in unwanted side effects.24 It was well reported that the volume of
the gallbladder would increase after oral administration of TGR5 agonists.13 To minimize the
2
ACS Paragon Plus Environment

Page 3 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
side effects, developing nonꢀsystemic TGR5 agonists specially targeting intestinal system
became our strategy.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Structures of some TGR5 agonists.
Since TGR5 is expressed in the intestinal tract11 and GLPꢀ1 is an intestinal peptide secreted
from colonic L cells as well,25 localized stimulation of TGR5 within the colon to initiate GLPꢀ1
release is possible.9 This hypothesis is also supported by an ex vivo explant study that TGR5
agonists induces GLPꢀ1 secretion in the explanted intestinal tract.26 Therefore, for the TGR5
agonists, the absorption from the intestine to the plasma may be not necessary to activate
intestinal TGR5 receptors, stimulate secretion of GLPꢀ1, and provide the desired antidiabetic
effect. Nonꢀabsorbed molecular therapeutics are usually developed to interact with intestinal
targets selectively, minimize offꢀtarget systemic effects, and reduce toxicity and sideꢀeffects.27
There are several other nonꢀabsorbed antidiabetic agents including the αꢀglucosidase inhibitor
acarbose,28 the apical sodiumꢀdependent bile acid transporter (ASBT) inhibitor 3ꢀ({[(3R,5R)ꢀ3ꢀ
butylꢀ3ꢀethylꢀ7ꢀ(methyloxy)ꢀ1,1ꢀdioxidoꢀ5ꢀphenylꢀ2,3,4,5ꢀtetrahydroꢀ1,4ꢀbenzothiazepinꢀ8ꢀ
yl]methyl}ꢀamino)pentanedioic acid (GSK2330672),29 and the nonꢀsystemic TGR5 agonists
developed by Exelixis, Inc.30 and Ardelyx, Inc..31 We focus our investigations on nonꢀabsorbed
TGR5 agonists with the expectation that these compounds would exert their therapeutic effects
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 44
1
2
3
4
locally in the gastrointestinal (GI) tract rather than in the gallbladder.
5
6
7
8
We sought to develop a series of nonꢀabsorbed TGR5 agonists by incorporating hydrophilic
groups on to the previously described series to reduce intestinal permeability and restrict
activation of TGR5 to the intestine.32 In our previous study of 4ꢀphenoxynicotinamide
derivatives as TGR5 agonists, a wide range of functional groups including chlorine, bromine,
methyl, and ethyl were well tolerated at 4ꢀposition of the phenoxy moiety, suggesting that this
position may fit into a large pocket within the TGR5 receptor (Figure 1).24 We hypothesized that
hydrophilic groups such as amino acids and PEG chains could be incorporated at this position to
minimize systemic exposure while maintaining potency.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The polar surface area (PSA) correlates well with passive molecular transport throuth
membranes and is an effective approach for predicting drug absorption in the GI tract (the Egan
Egg model).33 It incorporates most of Lipinski’s “rule of 5” for oral bioavailability, including
molecular weight, the number of hydrogen bond acceptors and donors within the compound.27,32
Drugs with a PSA < 60 Å2 are often completely absorbed from the intestine, while drugs with a
PSA > 140 Å2 are usually poorly absorbed.34,35 In this report, the calculated PSA was employed
as a predictive parameter to evaluate the gut permeability of TGR5 agonists, and our aim was to
synthesize analogues well outside the normal range for orally bioavailable drugs, e.g. > 140 Å2
PSA.
CHEMISTRY
Our routes for synthesizing amino acid or PEG containing derivatives of 2 are summarized in
Schemes 1ꢀ3. The key intermediate 4ꢀbromophenoxy derivative 6 was synthesized from the
commercially available 4ꢀchloronicotinic acid as previously reported by our group.24 As shown
in Scheme 1, 4ꢀchloronicotinic acid was converted into the acyl chloride by treatment with
4
ACS Paragon Plus Environment

Page 5 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
oxalyl chloride, followed by reaction with ethanol to afford the ester 3. Treatment of intermediate
3 with 4ꢀbromoꢀ2,5ꢀdichlorophenol in the presence of potassium carbonate yielded ester 4, and
subsequent hydrolysis of ester 4 provided acid 5. The aryl bromide intermediate 6 was obtained
by coupling of acid 5 with 1ꢀcyclopropylꢀ1,2,3,4ꢀtetrahydroquinoxaline. Heck reaction of aryl
bromide 6 with ethyl acrylate generated the desired intermediate 7.20 Reduction of the double
bond of olefin 7 with sodium borohydride in the presence of cuprous chloride gave compound 8.
A subsequent hydrolysis with sodium hydroxide yielded the acid 9. This acid was coupled with a
series of amino acid esters using HATU as a coupling reagent to produce the desired compounds
10a-c. Compouds 10d-f were obtained by hydrolysis of esters 10a-c. Compound 12 was
synthesized from intermediate 6 through a Heck reaction with isobutyl butꢀ3ꢀenoate followed by
hydrolysis with sodium hydroxide (Scheme 2). The miniꢀPEG analogues 15a-c were also
prepared from the intermediate 9 (Scheme 3). Coupling of acid 9 with amineꢀPEGꢀazide,
followed by Pd/Cꢀcatalyzed reduction of the azides with hydrogen led to the monoamine
derivatives 14a-c. After a second coupling reaction with acid 9 using HATU as a coupling
reagent, the monoamine derivatives 14a-c were converted into the corresponding dimers 15a-c.
Scheme 1. Synthesis of 10aꢀ10fa
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 44
1
2
3
4
5
6
7
8
Cl
Cl
Br
N
N
Cl
Cl
Cl
Cl
O
Br
Br
N
N
N
N
O
i
iv
ii
iii
N
O
Cl
COOH
Cl
O
COOEt
COOH
COOEt
9
3
4
5
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cl
Cl
Cl
Cl
Cl
COOEt
COOEt
COOH
N
N
N
N
N
O
O
O
v
vi
vii
Cl
O
N
O
O
N
N
N
7
9
8
10d: R2
10e: R2
10f: R2
=
=
N
H
COOH
10a: R1
10b: R1
10c: R1
=
=
=
N
COOEt
Cl
O
Cl
Cl
O
H
N
R1
N
N
R2
COOH
COOEt
N
H
N
H
O
O
ix
viii
Cl
=
N
O
O
N
COOH
N
N
COOEt
N
10a-c
10d-f
aReagents and conditions: (i) (a) SOCl2, reflux; (b) EtOH, Et3N, rt, 81%; (ii) 4ꢀbromoꢀ2,5ꢀ
dichlorophenol, K2CO3, DMF, 100 °C, 65%; (iii) NaOH, 1,4ꢀdioxane, H2O, rt; (iv) (a) (COCl)2,
DMF, DCM, reflux; (b) 1ꢀcyclopropylꢀ1,2,3,4ꢀtetrahydroquinoxaline, Et3N, DCM rt, 62% for
steps iii and iv; (v) ethyl acrylate, Pd(OAc)2, P(oꢀMePh)3, CH3CN, Microwave, 140 oC, 76%; (vi)
NaBH4, CuCl, THF, MeOH, 0ꢀ5 C, 82%; (vii) NaOH, 1,4ꢀdioxane, H2O, rt, 92%; (viii) HATU,
amino acid ester, Et3N, DCM, rt, 73ꢀ81%; (ix) NaOH, 1,4ꢀdioxane, H2O, rt, 85ꢀ90%.
o
Scheme 2. Synthesis of 12a
Cl
Cl
Cl
Cl
Cl
Cl
Br
N
N
N
N
COOiBu
ii
N
N
COOH
O
O
i
O
O
O
O
N
N
N
6
12
11
aReagents and conditions: (i) isobutyl butꢀ3ꢀenoate, Pd(OAc)2, P(oꢀMePh)3, CH3CN,
Microwave, 140 oC, 39%; (ii) NaOH, 1,4ꢀdioxane, H2O, rt, 86%.
Scheme 3. Synthesis of PEG Derivativesa
6
ACS Paragon Plus Environment

Page 7 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
Cl
Cl
O
Cl
Cl
Cl
Cl
O
O
COOH
O
N
N
N
N3
N
N
N
N
N
NH2
H
n
H
n
O
O
O
i
ii
O
O
O
13a: n = 3
13b: n = 6
13c: n = 8
6
14a: n = 3
14b: n = 6
14c: n = 8
N
N
N
7
8
13a-c
9
14a-c
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cl
Cl
O
O
Cl
Cl
O
N
N
N
N
H
N
15a: n = 3
15b: n = 6
15c: n = 8
n
H
O
O
O
iii
O
N
N
N
15a-c
aReagents and conditions: (i) HATU, Et3N, NH2ꢀPEGꢀN3, DCM, rt, 71%ꢀ78%; (ii) H2, Pd/C, rt;
(iii) 9, HATU, Et3N, DCM, rt, 49%ꢀ63% for steps ii and iii.
RESULTS AND DISCUSSION
A series of carboxylic acid functional groups was incorporated at the 4ꢀposition of the phenoxy
moiety (Table 1), and our goal was that these carboxylic acids could couple with amino acids or
miniꢀPEG chains. In vitro TGR5 assays demonstrated acceptable TGR5 potency for the
propanoic acid analogue 9 and marginal activity for the butenoic acid analogue 12. The
propanoic acid analogue 9 was therefore selected for further derivatization.
Table 1. In Vitro Activity of Compounds 8, 9 and 12.a
Cl
R
N
O
Cl
N
O
N
Compd
R
hTGR5 EC50 (nM)
mTGR5 EC50 (nM)
COOEt
3.8±1.0
11±2.0
8
9
COOH
35±1.3
71±7.9
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 44
1
2
3
4
378±86
405±72
12
COOH
5
6
7
aEC50 values given are expressed as the mean ± SD of three independent experiments.
8
9
Several amino acids were linked to 9 through an amide bond, and all of these derivatives
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
exhibited low nanomolar TGR5 activity and greater potency than the parent compound 9 (Table
2). In contrast to previously reported TGR5 agonists in which the mTGR5 potency was usually
much weaker than hTGR5 potency, these derivatives expressed mTGR5 potency comparable to
hTGR5 potency. Of note, compounds 10d, 10e, and 10f exhibited greater potency toward
mTGR5 than hTGR5. Introduction of amino acids greatly increased the PSA value, with most of
the derivatives possessing PSA values of greater than 100 Å2. A maximum PSA value of 112 Å2
was observed for compound 10d, and this compound was chosen for further in vivo
pharmacokinetic studies.
Table 2. In Vitro Activity of Compounds 10a−f.a
Compd
R
hTGR5 EC50 (nM)
mTGR5 EC50 (nM)
PSA (Å2)b
101
N
H
COOEt
10a
5.3±1.9
5.7±1.4
COOEt
N
H
12±3.9
5.0±2.2
17±2.6
7.8±2.4
101
92
10b
10c
N
COOEt
8
ACS Paragon Plus Environment

Page 9 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
N
H
COOH
112
112
103
10d
10e
10f
9.6±1.1
8.0±2.4
15±4.8
8.0±3.1
COOH
N
H
3.3±0.90
5.8±2.6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
N
COOH
aEC50 values given are expressed as the mean ± SD of three independent experiments. PSA
values were calculated using MarvinSketch 5.2.5.1.
b
Subsequent pharmacokinetic testing in rats indicated that 10d exhibited a low but could not be
neglected plasma exposure, with a Cmax of 13.5 ng/mL after a 5 mg/kg oral dose (Table 3). In
order to evaluate the in vivo glucoseꢀlowering activity of 10d, an oral glucose tolerance test
(OGTT) was carried out using ICR mice. Following oral administration of 10d (50 mg/kg), the
area under the glucose levels versus time curve (AUC) 0ꢀ120min was decreased by 33% compared
to the control group (Figure 2A and 2B). A further in vivo study in ICR mice revealed an 82%
increase in GLPꢀ1 concentration following administration of 10d compared to the control group
(Figure 2C). Combined treatment with 10d and Linagliptin (BIꢀ1356, (R)ꢀ8ꢀ(3ꢀaminoꢀpiperidinꢀ
1ꢀyl)ꢀ7ꢀbutꢀ2ꢀynylꢀ3ꢀmethylꢀ1ꢀ(4ꢀmethylꢀquinazolinꢀ2ꢀylmethyl)ꢀ3,7ꢀdihydroꢀpurineꢀ2,6ꢀdione),
a DPPꢀ4 inhibitor developed by Boehringer Ingelheim,36 increased secretion of GLPꢀ1 by 64%
compared to Linagliptin alone (Figure 2D). Unfortunately, gallbladder area increased
approximately 225% relative to the control group (Figure 3).
Table 3. PK properties of 10d, 15c, and the metabolite 9 following administration of 10d or 15c
to rats.a
Tmax
(h)
0.44
3.50
ꢀd
Cmax
(ng/mL)
13.5
15.5
AUC0ꢀt
(ng*h/mL)
AUC0ꢀ∞
(ng*h/mL)
MRT
(h)
2.37
5.47
ꢀ
T1/2
(h)
1.56
1.77
ꢀ
Compd
23.9
84.9
ꢀ
ꢀ
39.7
241
ꢀ
ꢀ
10d
9b
ꢀ
ꢀ
15c
9c
ꢀ
ꢀ
ꢀ
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 44
1
2
an = 4 rats/group; male SD rats (BK, Shanghai, China); p.o., 5 mg/kg for 10d, 20 mg/kg for 15c;
3
4
5
6
7
8
b
formulated in 0.25% CMCNa; PK parameters were calculated by WinNonlin 5.3. Data for
metabolite 9 following administration of 10d. cData for metabolite 9 following administration of
15c. d‘ꢀ’ indicates ‘no desired compound was detected’.
9
(A)
(B)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(C)
(D)
Figure 2. OGTT and GLPꢀ1 secretion study of 10d (50 mg/kg) in ICR mice: (A) blood glucose
concentration after administration of 10d; (B) blood glucose AUC0−120 min after administration of
10d. For (A) and (B): Compound 10d or 0.25% CMC (control) were administered orally to ICR
mice (n = 8 animals/group) at 0.5 h prior to oral glucose load (4 g/kg). Blood glucose levels were
measured before and after glucose load. (C) GLPꢀ1 secretion study of 10d. Compound 10d or
10
ACS Paragon Plus Environment

Page 11 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
0.25% CMC (control) were orally administered to ICR mice (n = 10 animals/group) 0.5 h prior
to oral glucose load (4 g/kg). Blood samples were collected 5 min after glucose challenge. (D)
GLPꢀ1 secretion study of 10d combined with Linagliptin. Linagliptin was orally administered to
ICR mice (n = 8ꢀ10 animals/group) 1 hour prior to compound 10d or 0.25% CMC (control)
dosing. 4 Hours after compounds or control administration, blood samples were collected. All
the blood samples were placed in Eppendorf tubes containing DPPꢀ4 inhibitor with final
concentrations of 1% blood samples and 25 mg/mL EDTA to measure serum active GLPꢀ1[7–36
amide] levels. ∗, P < 0.05; ∗∗, P <0.01 versus control. Error bar indicates SEM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Relative area of gallbladder after 10d (50 mg/kg) or 0.25% CMC (vehicle) was orally
administered to ICR mice (n = 8 animals/group). Following the OGTT experiment, the fasting
mice were refed for 2 h. The gallbladders were removed, and the area was measured using a
vernier caliper. The area of the gallbladder was measured by the length of the gallbladder
multiplied by the width of the gallbladder. ∗∗, P < 0.01 versus control. Error bar indicates SEM.
(A)
(B)
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 44
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(C)
(D)
Figure 4. OGTT and GLPꢀ1 secretion study following administration of compound 9 (50 mg/kg)
in ICR mice: (A) blood glucose concentration; (B) blood glucose AUC0−120 min. (C) GLPꢀ1
secretion. (D) GLPꢀ1 secretion study after administration of 9 combined with Linagliptin. ∗, P <
0.05; ∗∗, P <0.01 versus control. Error bar indicates SEM.
We suspected that the severe side effect occurring after administration of 10d was due to the
combined effects of 10d and its pharmacologically active metabolites. Pharmacokinetic studies
(0ꢀ24 h) revealed that the main metabolite (compound 9, mTGR5 EC50 = 71 nM) was formed by
hydrolysis of the amide bond of the parent compound. The metabolite 9 was observed in the
12
ACS Paragon Plus Environment

Page 13 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
plasma with a similar Cmax (15.5 ng/mL), but a higher AUC compared to the parent drug (Table 3,
Figure S1), and was demonstrated to be an active metabolite capable of activating TGR5 in vivo
and lowering the blood glucose levels (Figure 4A and 4B). After oral administration of
compound 9 (50 mg/kg), GLPꢀ1 levels increased by 57% compared to the control group (Figure
4C). When used in combination with Linagliptin, compound 9 also increased secretion of GLPꢀ1
by 55% compared to Linagliptin alone (Figure 4D). The combined activation of TGR5 by the
parent compound and active metabolite may lower the glucose concentrations and cause the
increase in gallbladder area.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. In Vitro Activity of Compounds 15a−c.a
Cl
O
O
Cl
Cl
O
N
N
N
N
N
N
n
H
H
O
O
Cl
O
O
N
N
`
Compd
15a
n
3
6
8
MW
1181
1313
1401
PSA (Å2)b
177
hTGR5 EC50 (nM)
30±13
mTGR5 EC50 (nM)
14±1.0
205
20±8.2
13±1.4
15b
223
25±1.7
12±2.9
15c
b
aEC50 values are expressed as the mean ± SD of three independent experiments. PSA values
were calculated using MarvinSketch 5.2.5.1.
Table 5. Apparent permeability in Cacoꢀ2 cells.a
A to B
B to A
Compd
Efflux Ratio
0.8
Papp
Papp
Recovery (%)
Recovery (%)
（10ꢀ6 cm/s）
6.75±1.61
（10ꢀ6 cm/s）
5.58±0.58
2
46.8±1.7
69.4±0.7
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 44
1
2
3
4
5
6
7
8
9
0.04±0.01##
0.1±0.03
99.5±4.2
2.24±0.19**, ##
15.8±4.5
91.2±1.5
95.8±0.3
61
15c
153
Digoxin
100.8±0.5
a‘A to B’ indicates the experiment from apical to basolateral, ‘B to A’ indicates the experiment
from basolateral to apical. ∗∗, P <0.01 versus Papp (A to B) of 15c; ##, P <0.01 versus Papp of 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Considering that 10d may still be absorbed to some extent, compounds with a much higher
PSA and lower permeability may be necessary to avoid activation of TGR5 receptors in the
gallbladder. A series of dimeric TGR5 agonists 15a-c containing a miniꢀPEG linker was
designed and synthesized (Table 4). The molecular weights of these analogues were greater than
1000, and we expected them to be poorly absorbed from the intestine based on Lipinski’s rule.33
The PSA values of the PEG3 derivative 15a, the PEG6 derivative 15b, and the PEG8 derivative
15c were 177, 205, and 223 Å2, respectively. Although introduction of PEG chains to dimeric
TGR5 represented a substantial departure from the original structure, the potency did not
decrease. The mTGR5 EC50 values for 15a-c were 14, 13, and 12 nM, and the hTGR5 EC50 for
15a-c were 30, 20, and 25 nM. The length of the PEG linker did not have a large effect on the
potency. The PEG8 derivative 15c was selected for further study due to its higher molecular
weight and greater PSA. In Cacoꢀ2 permeability assays, compound 15c exhibited a much lower
permeability than the smallꢀmolecule TGR5 agonist 2 (Table 5).
To determine the systemic exposure of 15c, a pharmacokinetic study was carried out (Table 3).
Following administration of an oral dose of 15c (20 mg/kg) to rats, blood samples were obtained
and analyzed (15 min ꢀ 24 h). There were no detectable (LLOQ = 3 ng/mL) levels of the parent
compound 15c, the potential metabolite 9, or any other major metabolites in the systemic
circulation.
14
ACS Paragon Plus Environment

Page 15 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. GLPꢀ1 secretion by NCIꢀH716 cells in response to a 2ꢀh treatment with PMA (1 ꢁM,
positive control), vehicle (control), or 15c (10 , 3, 1, and 0.5 ꢁM). Data shown are expressed as
the mean ± SEM of three independent experiments. ∗∗, P < 0.01 versus control.
Next, we examined the in vivo glucoseꢀlowering activity of 15c using an OGTT in ICR mice.
Considering the high molecular weight, the oral dose of 15c was increased to 150 mg/kg (0.11
mmol/kg). As shown in Figure 6A and 6B, the (AUC)0ꢀ120 min for glucose levels versus time
decreased 30% compared to the control group. Following a single dose, 15c significantly
lowered glycemia at time points of 15 and 30 min. Compound 15c exhibited a comparable
glucoseꢀlowering effect with absorbed TGR5 agonist 2.
We also evaluated the GLPꢀ1 secretion effect of 15c. Treatment of NCIꢀH716 cells with 15c
at 0.5, 1, 3, and 10 ꢁM concentrations, GLPꢀ1 secretion increased in a doseꢀdependent manner
(Figure 5). A further in vivo study in ICR mice was also carried out. After a single oral
administration, the plasma active GLPꢀ1 level was increased by 97% compared to the control
group (Figure 6C). Besides, compound 15c also increased basal secretion of GLPꢀ1. The
combination of 15c and Linagliptin enhanced GLPꢀ1 secretion by 61% compared to Linagliptin
alone (Figure 6D).
Following oral administration of 15c (150 mg/kg) to ICR mice, the gallbladder area increased
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 44
1
2
3
4
5
6
7
8
9
approximately 44% compared with the control group, which was not a significant difference
(Figure 7). This is in sharp contrast to compound 2, which increased gallbladder area severalꢀfold
after an oral dose of 50 mg/kg.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(A)
(B)
(C)
(D)
Figure 6. OGTT of 15c (150 mg/kg) and 2 (50 mg/kg) and GLPꢀ1 secretion study of 15c (150
mg/kg) in ICR mice: (A) blood glucose of 15c and 2; (B) blood glucose AUC0−120 min of 15c and
2. (C) GLPꢀ1 secretion study of 15c. Compound 15c or 0.25% CMC (control) was orally
administered to ICR mice (n = 10 animals/group) 1.5 h prior to the oral glucose load (4 g/kg). (D)
16
ACS Paragon Plus Environment

Page 17 of 44
Journal of Medicinal Chemistry
1
2
3
4
GLPꢀ1 secretion study of 15c combined with Linagliptin. ∗, P < 0.05; ∗∗, P <0.01 versus control.
5
6
Error bar indicates SEM.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Relative area of gallbladder after compound or 0.25% CMC (control) was orally
administered to ICR mice (n = 8 mice/group). Compound dose: 150 mg for 15c, 50 mg/kg for 2.
The fasting mice were refed for 2 h after the OGTT experiment, then, the gallbladders were
removed, and the area was measured using a vernier caliper. The relative area of the gallbladder
was calculated from the length multiplied by the width of the gallbladder. ∗∗, P < 0.01 versus
control. Error bar indicates SEM. Data are representative of three independent experiments.
We further analyzed the drug levels of 15c in the gallbladder. As depicted in Figure 8, four
hours after oral administration of 15c (150 mg/kg) there was no parent drug present in the
gallbladder tissue (< 0.3 ng) and the metabolite 9 was only present at a low concentration. The
bile concentrations of 15c and 9 were far lower than the concentration of 2 after oral
administration of 50 mg/kg.
In another PK experiment, we treated ICR mice with the same dosing as used in the OGTT
experiment, 150 mg/kg p.o. of 15c, and measured the drug levels in plasma, bile, and gallbladder
tissue at time 2, 2.5, and 3.5 h after administration. As shown in Table 6, only nanomolar levels
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 44
1
2
3
4
5
6
7
8
9
of 15c were detected in the plasma and the bile, below the EC50 of 15c (mTGR5 EC50 = 12 nM).
The low concentrations of 15c in plasma, bile, and gallbladder tissue demonstrated that 15c was
a lowꢀabsorbed TGR5 agonist. The glucose lowering effect was due to its specific activation of
TGR5 in the intestine, and this maybe the reason for the decreased gallbladder side effects.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(A)
(B)
Figure 8. Drug levels of compound 15c (150 mg/kg), 9 (metabolite after orally dosed 15c) and 2
(50 mg/kg) in gallbladder 4 h after oral administration to ICR mice. (A) Drug levels in bile; (B)
Total drug levels in gallbladder tissue. n = 4−7 mice/group. Error bar indicates SEM.
Table 6. Drug levels of compound 15c in plasma, bile, and gallbladder tissue after oral
administration to ICR mice.a
Sample location
Time (h)
Plasma (ng/mL)
6.22 (4.4 nM)
9.65 (6.9 nM)
4.03 (2.9 nM)
Bile (ng/mL)
5.07 (3.6 nM)
4.93 (3.5 nM)
5.33 (3.8 nM)
Gallbladder tissue (ng/g)
2
ꢀb
ꢀ
2.5
3.5
ꢀ
aDrug levels of compound 15c in plasma, bile and gallbladder tissue after oral administration to
ICR mice (150 mg/kg). (A) Drug levels in bile; (B) Total drug levels in gallbladder tissue. n = 3
18
ACS Paragon Plus Environment

Page 19 of 44
Journal of Medicinal Chemistry
1
2
mice/group. b‘-’ indicates ‘no desired compound was detected’.
3
4
5
6
7
8
9
The appearance of 15c in the gallbladder indicated that there was still some parent compound
absorbed from the intestine, and an even longer PEG linker may be beneficial to further decrease
the absorption. Considering the active metabolite 9 originates from hydrolysis of the amide bond
to the PEG, it may be beneficial to introduce bulky groups to the amide to increase steric
hindrance and improve resistance to hydrolysis, or to replace the amide bond with an inherently
more resistant ether or amine linkage. The effectiveness of the miniꢀPEG linkage suggested that
there may be a large pocket available on the TGR5 receptor, and that other hydrophilic groups,
such as quaternary ammonium salts or polyhydric derivatives could also be incorporated into the
4‑phenoxynicotinamide scaffold to prepare some nonꢀabsorbed TGR5 agonists while
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
maintaining the potency.
Conclusion
In summary, introduction of polar, hydrophilic groups to the 4ꢀphenoxynicotinamide
derivative 2 as well as dimerizing the core structure using a PEGꢀlinker led to a series of lowꢀ
absorbed TGR5 agonists. Due to the low intestinal absorption, compound 15c, which displays in
vivo glucose lowering effects comparable to the absorbed TGR5 agonist 2, reduced the
gallbladder side effect a lot compared to 2. As the first lowꢀabsorbed TGR5 agonist to be
reported, 15c represents a useful prototype compound in the development of other nonꢀabsorbed
TGR5 agonists for diabetes treatment.
EXPERIMENTAL SECTION
In Vitro TGR5 Assay.
hTGR5/CRE/HEK293 or mTGR5/CRE/HEK293 stable cell line was obtained by transfection
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 44
1
2
3
4
5
6
7
8
9
of HEK293 cells with human or mouse TGR5 expression plasmid (hTGR5ꢀpcDNA3.1 or
mTGR5ꢀpcDNA3.1) and CREꢀdriven luciferase reporter plasmid (pGL4.29, Promega, Madison,
WI), and employed to assess the activity of test compounds by reporter gene assay. Briefly, cells
were seeded into 96ꢀwell plates and incubated overnight in DMEM supplemented with 10% FBS
in 5% CO2 at 37 °C. Then, cells were incubated with fresh medium containing different
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
concentrations of test compounds or 20 µM compound 1 as a positive control for 5.5 hours.
Luciferase activity in cell lysate was determined using the SteadyꢀGlo® Luciferase Assay
System (Promega, Madison, WI) according to the manufacturer’s instructions. Assays were
performed for at least 3 times.
In Vitro GLP-1 Assay.
NCIꢀH716 cells were maintained in suspension culture as described by American Type Culture
Collection. Fortyꢀeight hours before the assay, cells were planted into 24ꢀwell plates coated with
Matrigel (BD Biosciences, Oxford, UK). On the assay day, cells were washed with KREBS and
then incubated with test reagents diluted in KREBS (with 0.2% BSA and 1% DPPꢀ4 inhibitor)
for 2 h at 37 °C. After incubation, buffer was collected and centrifuged to remove floating cells.
Concentrations of GLPꢀ1 in the buffer were assayed using active GLPꢀ1 ELISAꢀkit from
Millipore.37,38 Results were mean of three independent assays in duplicate.
Animals.
Male ICR mice (7ꢀ8 weeks) were purchased from the Shanghai SLAC Laboratory Animal Co.
Ltd (Shanghai, China). The animals were maintained under a 12ꢀh lightꢀdark cycle with free
access to water and food. Animal experiments were approved by the Animal Care and Use
Committee, Shanghai Institute of Materia Medica, Chinese Academy of Sciences.
GLP-1 Secretion in ICR Mice.
20
ACS Paragon Plus Environment

Page 21 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Glucose stimulated GLPꢀ1 secretion: compounds 9 (50 mg/kg), 10d (50 mg/kg) and 15c (150
mg/kg) were orally administered to overnight fasting male ICR mice (n = 9ꢀ10 animals/group, 7ꢀ
8 weeks). Compound 15c was given 1.5 hours prior to glucose challenge. Compound 9 and 10d
were given 0.5 hour prior to the glucose challenge. Then all the mice were challenged with 4
g/kg glucose, and blood samples were collected at 5 mins after glucose loading.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Basal GLPꢀ1 secretion: male ICR mice were fasted for 6 hours. Linagliptin (3 mg/kg) was
orally given to all the mice except vehicle control group. 1 Hour later, compound 9 (50 mg/kg),
10d (50 mg/kg) and 15c (150 mg/kg) were orally given to the mice. Blood samples were
collected 4 hours after compounds dosing.
All the blood samples were placed in Eppendorf tubes containing the DPPꢀ4 inhibitor
(Millipore, DPPꢀ4ꢀ010) with a final concentration of 1% blood samples and 25 mg/mL EDTA to
measure serum active GLPꢀ1[7ꢀ36 amide] levels. All the compounds were evaluated in one
examination, so Figure 2C, 4C, 6C share the same control, and Figure 2D, 4D, 6D share the
same control
Oral Glucose Tolerance Test (OGTT) and Gallbladder Area Measurement in ICR Mice.
To examine the acute effect of compounds on blood glucose after an oral glucose challenge,
test compounds or vehicle (0.25% CMC) were orally administered to overnight fasting male ICR
mice (n = 8 mice/group, 7ꢀ8 weeks) 0.5 (10d and 9) or 1.5 (15c) h prior to the oral glucose load
(4 g/kg). Blood glucose levels were measured via blood drops obtained by clipping the tail of the
mice using an ACCUꢀCHEK Advantage II Glucose Monitor (Roche, Indianapolis, IN) before
compound dosing and 0, 15, 30, 60 and 120 min after the glucose load. The area under the
concentration–time curve from 0 to 120 min (AUC0ꢀ120min, Glu) of blood glucose after the glucose
load was calculated by the trapezoidal rule (using the 0 time glucose level for each animal as that
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 44
1
2
3
4
5
6
animal’s baseline). After the OGTT experiment, the fasting mice were reꢀfed for 2 h. Then,
gallbladders were removed, and the area was measured using vernier caliper.
7
8
9
PK experiment in SD rats.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Male SD rats (200ꢀ220 g) were obtained from Shanghai SLAC Laboratory Animal Co. Ltd
(Shanghai, China). Compound 10d (5 mg/kg), or 15c (20 mg/kg) was orally administered to 12ꢀ
fasting SD rats (n = 4). Blood samples were collected at different time points. After
centrifugation (11000 rpm, 5 min) of the blood, plasma was isolated. 25 ꢁL of the sample was
mixed with 12.5 ꢁL of methanol, 12.5 ꢁL of H2O and 25 ꢁL of phenacetin solution (1000 ng/mL).
After centrifugation at 11000 rpm for 5 mins, the supernatant was analyzed by LCꢀMS/MS. We
tested the drug levels of different time points from 0.25 h to 24 h, and we chose one of the time
points with the highest drug concentration as the Cmax and Tmax (Figure S1). The other parameters
were calculated based on the drugꢀtime curve by WinNonlin 5.3.
Caco-2 permeability determination
Cacoꢀ2 cells were obtained from the ATCC (HTBꢀ37) and maintained in DMEM containing
10% FBS, 1% glutamine, 1% nonessential amino acids, 100 mg/mL penicillin and streptomycin.
Cacoꢀ2 cells were cultured at 37 oC in a 5% CO2 and 90% relative humidity environment. Cells
were used between passages 30 and 40. After 21 days of culture, the integrity of the cell
monolayer was verified by measuring the transepithelial electrical resistance (TEER). Drug
transport from the apical side to the basolateral side (AꢀB) and from the basolateral side to the
apical side was measured under the same condition. To verify our model is proper for both
compounds with low and high permeabilities, propranolol (high permeability) and atenolol (poor
permeability) were used as the external hypertonic and hypotonic control, respectively. For the
experiment, the Papp values of propranolol are 19.2 × 10ꢀ6 cm/s (A to B) and 12.9 × 10ꢀ6 cm/s (B
22
ACS Paragon Plus Environment

Page 23 of 44
Journal of Medicinal Chemistry
1
2
to A). The Papp values of propranolol are 0.1 × 10ꢀ6 cm/s (A to B) and 0.6 × 10ꢀ6 cm/s (B to A).
The data indicates that this model works well for both high and low permeable compounds.
Digoxin was used as the positive control for Pgpꢀmediated drug efflux. The concentrations for
the tested compounds are 20 µM. In brief, the method was as follows. After washing the
monolayer with HBSS three times, the compounds were diluted and added to the appropriate
well (pH 6.8 for apical side and pH 7.4 for basolateral side). The plate was incubated at 37 oC for
95 min. Samples were collected from the donor side at 5 and 95 min, and from the receiver side
at 35 and 95 min postꢀincubation. The concentration of samples was measured by LCꢀMS/MS.
The Papp was calculated from the following equation: Papp = (VA/(SA×T))×
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
([drug]acceptor/[drug]initial donor). Where VA is the volume of the acceptor well, SA is the surface
area of the membrane, T is the total transport time, [drug]acceptor is the drug level at the acceptor
side, and [drug]initial donor is the drug level at the donor side at T = 0.
Drug Levels Test in Bile, and Gallbladder Tissue after Oral Administration to ICR Mice.
Compound 15c (150 mg/kg) or 2 (50 mg/kg) was administered orally to male ICR mice (7ꢀ8
weeks). Bile and gallbladder tissue were collected at different time points postꢀdose for drug
levels test. Briefly, gallbladder was homogenized after adding 100 µL of methanol. A 15 µL
aliquot of bile samples were diluted with 100 µL of methanol and then mixed. Bile and tissue
homogenate were stored at −20 °C. These samples were protein precipitated with acetonitrile
using clopidogrel as the internal standard, and the supernatant was injected onto LCMS/MS
system for the quantification. Chromatographic separation was performed on a luna C18 column
(50 mm × 2.0 mm ID, 5 µm, Phenomenex, Torrance, CA, USA) using 5 mM ammonium acetate
and acetonitrile (1: 1, v/v) containing 0.02% formic acid as the mobile phase, which was
delivered at a flow rate of 0.8 mL/min. The MS detection was carried out in multiple reactions
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 44
1
2
3
4
5
6
7
8
9
monitoring mode using a positive electrospray ionization interface. The calibration curve of
compound was established from 3.0 to 3000 ng/mL for bile, and from 3.0 to 3000 ng/g for
gallbladder tissue. To confirm accuracy, quality control samples were analyzed in duplicate at
three different concentrations. The bioanalytical run meet the acceptance criteria which require
that at least 3/4 of the calibration standards were within 15% of theoretical and that greater than
2/3 of all quality control samples were within 15% of theoretical. All the equipments and
parameters for the drug levels test and PK experiment were list in supporting information.
Statistical Analysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
All data were expressed as the Mean ± SEM. For the experiment including multiple time
points, oneꢀway ANOVA statistical analysis was used, otherwise, unpaired Student’s t test was
used.
Synthetic Materials and Methods.
All reagents were purchased from commercial suppliers and used without further purification.
13
1H NMR and C NMR spectra were recorded on a Brucker AC300 or a Brucker AC400 NMR
spectrometer using tetramethylsilane as an internal reference. ESIꢀMS spectra were obtained on a
Krats MS 80 mass spectrometer. The purity of tested compounds was determined by HPLC
(Agilent LC1200, Agilent ChemStation, Agilent Eclipse XDBꢀC18, 5 ꢁM, 4.6×150 mm, UV 254
nM, 30 oC, Flow Rate = 1.0 mL/min). All the assayed compounds possess > 95% purity. Column
chromatography was performed on silica gel (200ꢀ300 mesh).
Ethyl 4-Chloronicotinate (3). 4ꢀChloronicotinic acid (3.0 g, 19.1 mmol) was dissolved in
thionyl chloride (30 mL), and the solution was refluxed for 1.5 h. Then the solvent was removed
under reduced pressure. The residue was dissolved in toluene (5 mL) and concentrated to dryness.
Then ethanol (30 mL) and triethylamine (5 mL) were added to the residue, and the solution
24
ACS Paragon Plus Environment

Page 25 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mixture was stirred at room temperature overnight. After the solvent was evaporated, H2O (30
mL) was added to the residue, and then extracted with dichloromethane (30 mL × 3). The
combined organic phase was washed with saturated solution of sodium bicarbonate (50 mL) and
brine (50 mL), and then dried over anhydrous sodium sulfate. The product was purified by flash
column chromatography (petroleum ether/ethyl acetate = 8:1) to yield 19 as colorless oil (0.95 g,
81%). LCꢀMS (ESI+): 186 (M + H)+. 1H NMR (300 MHz, CD3OD) δ 9.20 (s, 1H), 8.82 (d, J =
6.1 Hz, 1H), 8.08 (d, J = 6.1 Hz, 1H), 4.58 – 4.37 (m, 2H), 1.50 – 1.36 (m, 3H).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ethyl 4-(2,5-Dichloro-4-bromo-phenoxy)-nicotinate (4). To a solution of 3 (1.85 g, 10 mmol)
in anhydrous DMF (15 mL), 2,5ꢀdichloroꢀ4ꢀbromophenol ( 2.4 g, 12 mmol) and potassium
carbonate (4.14 g, 30 mmol) were added. After the solution was heated at 100 °C for 3 h, the
reaction was cooled to room temperature and poured into H2O (100 mL). The mixture was
extracted with ethyl acetate (100 mL × 3), and the combined organic phase was washed with
brine. After dried over anhydrous sodium sulfate, the solvent was evaporated to afford the
product. The crude was purified with flash column chromatography (petroleum/ethyl acetate =
1
4:1) to yield the desired compound in 65% yield. H NMR (400 MHz, CDCl3) δ 9.09 (s, 1H),
8.57 (d, J = 5.7 Hz, 1H), 7.26 (s, 1H), 7.22 (d, J = 4.3 Hz, 1H), 6.62 (d, J = 5.8 Hz, 1H), 4.40 (q,
J = 7.1 Hz, 2H), 1.37 (t, J = 7.1 Hz, 3H).
[4-(2,5-Dichloro-4-bromophenoxy)-pyridin-3-yl]-(4-cyclopropyl-3,4-dihydro-2H-
quinoxalin-1-yl)-methanone (6). To a solution of 4 (2.4 g, 6.13 mmol) in 1,4ꢀdioxane (20 mL)
and H2O (20 mL), sodium hydroxide (0.49 g, 6.13 mmol) was added. The reaction was stirred at
ambient temperature for 3 h. After the solvent was evaporated, the residue was dissolved in H2O
and pH was adjusted to 3 with 2N hydrochloric acid. The formed precipitated was filtered,
washed with cold water and dried under vacuum providing 5. The obtained compound was
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 44
1
2
3
4
5
6
7
8
9
dissolved in dichloromethane (30 mL), and then oxalyl chloride (1.56 mL, 18.4 mmol) was
added to the solution. The resulting solution was refluxed for 2 h. After the reaction was
completed, the solvent was evaporated and the remaining oxalyl chloride was removed by
azeotropic distillation with toluene. After the residue was dissolved in anhydrous
dichloromethane, 1ꢀcyclopropylꢀ1,2,3,4ꢀtetrahydroꢀ quinoxaline (1.07 g, 6.13 mmol) and Et3N
(2.55 mL, 18.39 mmol) were added to this solution. And the reaction was stirred at room
temperature for 2 h. After the reaction was completed, the mixture was washed with brine, dried
over anhydrous sodium sulfate and purified by flash column chromatography (petroleum/ethyl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
acetate = 3:1) to provide 6 (1.97 g) in 62% yield for two steps. H NMR (300 MHz, CDCl3): δ
8.85 (s, 1H), 8.44 (d, J = 4.2 Hz, 1H), 7.67 (s, 1H), 7.02 (t, J = 5.7 Hz, 1H), 6.94 (d, J = 5.7 Hz,
1H), 6.47 (d, J = 5.7 Hz, 1H), 6.38 (t, J = 5.7 Hz, 1H), 6.30 (d, J = 4.2 Hz, 1H), 5.76 (s, 1H),
4.89 (m, 1H), 3.49 (m, 2H), 3.16 (m, 1H), 2.25 (m, 1H), 0.69 (m, 3H), ꢀ0.23 (br, 1H).
(E)-ethyl
3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-ca-
rbonyl)pyridin-4-yl)oxy)phenyl)acrylate (7). To a suspension of 6 (1.04 g, 2.00 mmol) in
acetonitrile (15 mL), ethyl acrylate (534 ꢁL, 4.00 mmol), palladium diacetate (89.8 mg, 0.40
mmol), tris(2ꢀmethylphenyl)phosphine (243 mg, 0.80 mmol) and triethylamine (1.39 mL, 10.0
mmol) were added. The resulting mixture was heated using microwave reactor at 140 °C for 1 h.
After cooling to ambient temperature, the reaction mixture was filtered, and the filtrate was
concentrated under reduced pressure. The residue was purified by flash column chromatography
(petroleum ether/ethyl acetate = 4:1) to afford the title compound 7 (0.82 g, 76%) as a light
1
yellow foam. H NMR (300 MHz, CDCl3): δ 8.88(s, 1H), 8.46(d, J = 5.4 Hz, 1H), 7.91(d, J =
15.6 Hz, 1H), 7.65(s, 1H), 7.04ꢀ6.99(m, 1H), 6.89(d, J = 8.1 Hz, 1H), 6.48(d, J = 8.7 Hz, 1H),
6.42ꢀ6.37(m, 3H), 5.73(s, 1H), 4.89(m, 1H), 4.28(q, J = 3.9 Hz, 2H), 3.49(m, 2H), 3.17(m, 1H),
26
ACS Paragon Plus Environment

Page 27 of 44
Journal of Medicinal Chemistry
1
2
3
4
2.22(m, 1H), 1.35(t, J = 3.9 Hz, 3H ), 0.65(m, 3H), ꢀ0.27(m, 1H).
5
6
Ethyl
3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-carb-
7
8
9
onyl)pyridin-4-yl)oxy)phenyl)propanoate (8). To a mixture of 7 (800 mg, 1.48 mmol) and
cuprous chloride (147 mg, 1.48 mmol) in methanol (7 mL) and tetrahydrofuran (7 mL), NaBH4
(253 mg, 6.66 mmol) was added at 0 °C, and the resulting mixture was stirred at 0ꢀ5 °C for 2 h.
Then the reaction was quenched by the addition of water (2 mL). The resulting solid was filtered
off, and the filtrate was concentrated under reduced pressure. The residue was diluted with water
(10 mL), extracted with ethyl acetate (10 mL × 2). The organic layers were combined, dried over
anhydrous magnesium sulfate and evaporated under reduced pressure. The residue was purified
by flash column chromatography (petroleum ether/ethyl acetate = 3:1) to give 8 (655 mg, 82%)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
as a light yellow slurry. H NMR (300 MHz, DMSO): δ 8.82(s, 1H), 8.40(d, J = 4.2 Hz, 1H),
7.30(s, 1H), 7.03(m, 1H), 6.93(d, J = 6.3 Hz, 1H), 6.48(d, J = 6.0 Hz, 1H), 6.37(m, 1H), 6.29(d,
J = 4.2 Hz, 1H), 5.69(s, 1H), 4.89(m, 1H), 4.15(q, J = 5.4 Hz, 2H), 3.49(m, 2H), 3.17(m, 1H),
2.99(t, J = 5.7 Hz, 2H), 2.61(t, J = 5.7 Hz, 2H), 2.25(m, 1H), 1.26(t, J = 5.4 Hz, 3H), 0.62(m,
3H), ꢀ0.34(m, 1H). 13C NMR (126 MHz, CDCl3) δ 171.57, 163.28, 158.21, 151.81, 151.34,
147.06, 139.80, 136.25, 132.12, 131.08, 126.12, 124.73, 124.16, 123.00, 122.65, 121.88, 115.00,
112.63, 108.78, 66.59, 60.24, 48.37, 39.73, 33.13, 30.42, 27.67, 13.73, 7.18. HRMS (ES+) calcd
for C28H27Cl2N3O4Na (M+Na)+ m/z 562.1276, found 562.1272.
3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-carbonyl)pyridin-4-
yl)oxy)phenyl)propanoic acid (9). To a solution of 8 (639 mg, 1.18 mmol) in 1,4ꢀdioxane (5 mL)
and water (5 mL), sodium hydroxide (94.4 mg, 2.36 mmol) was added, and the resulting mixture
was stirred at ambient temperature for 3 h. The reaction was evaporated to dryness, and the
residue was dissolved in water (10 mL). The pH of the solution was adjusted to 3 with 2N HCl.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 44
1
2
3
4
The formed precipitate was filtered, washed with cold water, and dried under vacumm to provide
5
6
1
the title compound 9 (556 mg, 92%) as a light yellow solid. H NMR (300 MHz, DMSO): δ
7
8
9
12.29(s, 1H), 8.77(s, 1H), 8.41(d, J = 3.9 Hz, 1H), 7.59(s, 1H), 6.97(m, 1H), 6.84(m, 1H),
6.53(m, 1H), 6.48(m, 1H), 6.33(m, 1H), 5.54(s, 1H), 4.71(m, 1H), 3.36(m, 2H), 3.07(m, 1H),
2.86(m, 2H), 2.51(m, 1H), 2.16(m, 1H), 0.60(m, 2H), 0.47(m, 1H), ꢀ0.44(m, 1H). 13C NMR (126
MHz, CDCl3) δ 175.79, 163.03, 158.56, 151.23, 150.77, 146.91, 139.84, 136.48, 132.20, 131.11,
126.23, 124.61, 124.18, 122.94, 122.83, 121.93, 115.05, 112.70, 108.98, 48.35, 39.81, 33.15,
30.46, 27.71, 7.22. HRMS (ES+) calcd for C26H24Cl2N3O4 (M+H)+ m/z 512.1144, found 512.1141.
General procedure for 10a–c. To a solution of 9 (0.2 mmol) in dichloromethane (5 mL),
HATU (0.3 mmol), triethylamine (0.6 mmol) and amino acid ester (0.3 mmol) were added. The
resulting mixture was stirred at room temperature overnight. The reaction mixture was diluted
with water (10 mL) and extracted with dichloromethane (10 mL). The organic layer was washed
with saturated brine, dried over anhydrous magnesium sulfate, and concentrated under reduced
pressure. The residue was purified by flash chromatography (petroleum ether/ethyl acetate = 2:1
to 1:1) to give the desired compound.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ethyl
2-(3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-ca-
rbonyl)pyridin-4-yl)oxy)phenyl)propanamido)acetate (10a). The title compound was
obtained from 9 and glycine ethyl ester hydrochloride according to the general procedure in yield
1
of 81%. H NMR (300 MHz, CDCl3): δ 8.81(s, 1H), 8.40(m, 1H), 7.31(s, 1H), 7.03(m, 1H),
6.96(m, 1H), 6.48(m, 1H), 6.39ꢀ6.31(m, 2H), 5.98(t, J = 4.2 Hz, 1H), 5.72(s, 1H), 4.87(m, 1H),
4.20(q, J = 7.2 Hz, 2H), 4.02(m, 2H), 3.48(m, 2H), 3.20(m, 1H), 3.03(t, J = 7.5 Hz, 2H), 2.53(t, J
13
= 7.5 Hz, 2H), 2.26(m, 1H), 1.28(t, J = 7.2 Hz, 3H), 0.64(m, 3H), ꢀ0.28(m, 1H). C NMR (126
MHz, CDCl3) δ 172.37, 170.55, 163.19, 158.39, 151.60, 151.08, 146.93, 136.66, 132.05, 131.07,
28
ACS Paragon Plus Environment

Page 29 of 44
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
126.14, 124.69, 124.23, 122.97, 122.68, 121.93, 115.04, 112.64, 108.86, 60.35, 48.36, 39.76,
35.26, 34.37, 33.42, 30.45, 28.20, 13.68, 7.23. HRMS (ES+) calcd for C30H30Cl2N4O5Na
(M+Na)+ m/z 619.1491, found 619.1482.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ethyl
3-(3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-ca-
rbonyl)pyridin-4-yl)oxy)phenyl)propanamido)propanoate (10b). The title compound was
obtained from 9 and βꢀalanine ethyl ester hydrochloride according to the general procedure in
1
yield of 79%. H NMR (300 MHz, CDCl3): δ 8.81(s, 1H), 8.41(d, J = 6.0 Hz, 1H), 7.30(s, 1H),
7.03(m, 1H), 6.96(m, 1H), 6.46(m, 1H), 6.33(m, 1H), 6.31(m, 1H), 6.11(d, J = 6.0 Hz, 1H),
5.73(s, 1H), 4.85(m, 1H), 4.13(q, J = 6.9 Hz, 2H), 3.54ꢀ3.49(m, 4H), 3.21(m, 1H), 3.00(t, J = 7.5
Hz, 2H), 2.51(t, J = 6.0 Hz, 2H), 2.44(t, J = 7.5 Hz, 2H), 2.26(m, 1H), 1.25(t, J = 6.9 Hz, 3H),
0.65(m, 3H), ꢀ0.27(m, 1H).13C NMR (126 MHz, CDCl3) δ 170.69, 169.46, 163.25, 158.34,
151.68, 151.16, 147.04, 139.81, 136.43, 132.02, 131.21, 126.12, 124.72, 124.32, 122.97, 122.64,
121.97, 115.04, 112.64, 108.87, 61.20, 48.36, 40.87, 39.75, 34.93, 30.45, 28.12, 20.17, 13.64,
7.22. HRMS (ES+) calcd for C31H32Cl2N4O5Na (M+Na)+ m/z 633.1647, found 633.1651.
Ethyl
2-(3-(2,5-Dichloro-4-((3-(4-cyclopropyl-1,2,3,4-tetrahydroquinoxaline-1-ca-
rbonyl)pyridin-4-yl)oxy)phenyl)-N-methylpropanamido)acetate (10c). The title compound
was obtained from 9 and ethyl sarcosinate hydrochloride according to the general procedure in
1
yield of 73%. H NMR (300 MHz, CDCl3): δ 8.81(s, 1H), 8.40(d, J = 6.0 Hz, 1H), 7.35(s, 1H),
7.04ꢀ6.97(m, 2H), 6.49(m, 1H), 6.34(m, 2H), 5.74(s, 1H), 4.85(m, 1H), 4.19(q, J = 6.9 Hz, 2H),
4.12(s, 2H), 3.48(m, 2H), 3.07ꢀ2.99(m, 6H), 2.67(m, 2H), 2.26(m, 1H), 1.27(t, J = 6.9 Hz, 3H),
0.64(m, 3H), ꢀ0.24(m, 1H). 13C NMR (126 MHz, CDCl3) δ 171.47, 168.74, 163.28, 158.35,
151.68, 151.15, 146.91, 139.80, 137.00, 132.05, 131.42, 126.10, 124.71, 124.30, 122.98, 122.64,
121.89, 115.04, 112.63, 108.83, 60.76, 49.01, 48.35, 39.74, 35.97, 34.58, 32.07, 30.48, 27.81,
29
ACS Paragon Plus Environment